Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Economic Freedom (for the Rest of Us) by Jim Stanford Economist, Canadian Auto Workers Suite 1306 – 2000 Barrington St., Halifax, B3J 3K1 (902) 455-9327 fax 454-9473 email: [email protected] January 1999 What is Economic Freedom? All Canadians are equally free to live in a cardboard box under a freeway. Why is it, then, that only the most desperate and poor among us choose to exercise this “freedom”? When it comes to the economy, “freedom” is a tricky concept. Whether or not one is “free” depends very much on where one fits into the economic pecking order. The freedom to sleep under a freeway is not very relevant to a millionaire. Similarly, the freedom to invest one’s financial capital in foreign countries without government restriction is not very relevant to a homeless person. In comparison to fundamental political principles such as the right to hold free elections and the right to freedom of speech, there are fewer absolutes in trying to define what economic freedom means. But this difficulty has not stopped the proponents of conservative economic policies from arguing that their measures will enhance the economic freedom of society. The very names attached to conservative policies reflect this presumed oneness with greater “freedom”: free trade, the free market, free commerce, free competition. These ideologically loaded terms reflect a common and equally loaded underlying assumption—namely that a “free” market for businesses and investors translates into more freedom for all of us. Needless to say, this vision of freedom is applied to real-world policy debates in a selective, even slippery fashion. “Free Canadian savings,” cried a Globe and Mail editorial calling on the federal government to abolish the 20 percent ceiling on the foreign investment of RRSP monies.1 Contrary to the implication of the editorial, Canada has no controls on the outward export of capital whatsoever: any individual can transfer as much of their savings as they want out of this country to any other country that will accept it. The editorialists were actually complaining that the federal government refuses to subsidize RRSPs and other private pension funds (with tax deductions worth over 50 percent for high-income earners in most provinces) if more than one-fifth of those funds are invested abroad. The editorialists want both the government subsidy and the right to move that subsidized money abroad; it seems that some forms of government interference (the foreign content limit) are more problematic than others (the generous subsidy for RRSPs). Nevertheless, the image of unjustly imprisoned savings somehow languishing in a Canadian jail is a powerful one. And by invoking the “freedom” card, the Globe and Mail is simply following the well-trodden path of those who have advocated more pro-business, pro-investor policies for generations. For the advocates of conservative economic policies, the free market represents the epitome of economic freedom. Yet bitter economic experience demonstrates clearly that for the propertyless, the free market can be a prison. Countless struggles by generations of working people have been aimed at subjecting market forces to democratic regulation in the interests of equality, fairness, and opportunity: struggles to form unions and bargain over wages and working conditions, efforts to establish social policies which redistribute income and provide minimum 1 economic protections to the poorest members of society, attempts to hold private corporations more accountable through regulation or taxes to the societies from which they generate their profits. Through the lens of free-market economics, these efforts are intrusive and distortionary. But they nevertheless reflect the desire of masses of people to collectively better their economic prospects, and hence very much represent efforts to extend the principles of freedom and democracy to the economic sphere. Strangely, according to the mindset of a free-market ideologue, a society which decides collectively and democratically to impose limits on the powers of private markets becomes “un-free”; the exercise of political freedom is thus held to undermine economic freedom.2 Freedom for Whom? It was in the hallowed tradition of the “free” market that the Vancouver-based Fraser Institute established an ongoing project several years ago to attempt to measure and monitor the course of “economic freedom” in various jurisdictions around the world.3 This project assigned actual numerical scores to different jurisdictions on the basis of various indices of “economic freedom.” The variables considered include: the extent to which the value of financial wealth is protected (through such indicators as low inflation and positive real interest rates); the degree to which government intervenes in the economy through taxes, production, consumption, and economic regulation (such as price controls); and the extent to which international trade and investment relations are free from government interference. The ideological bias of the Fraser Institute’s approach to economic “freedom” is clear from the choice of variables considered by their index. An economy is “free” if the operations of private businesses and investors are relatively unfettered by government policies, rules, or practices which undermine the flexibility, profitability, or sustainability of those operations. Most people in the world do not own businesses or support themselves from the proceeds of financial wealth, but rather must attempt to support themselves through current work activity in an employment relationship—that is, they must get a job. But key economic policies or regulations which may make it easier or harder for workers to support themselves through work effort are not considered in the Fraser Institute’s ranking. The absence of indicators such as labour and trade union freedoms, employment policy, and social security programs from the Fraser Institute’s index speaks volumes about the one-sided nature of the “freedom” that this project endeavours to quantify. Since the Fraser Institute’s conception of economic freedom is so narrow, it is not surprising that their index sheds relatively little light on the actual economic conditions experienced by the majority of humanity—those that do not own businesses or significant accumulations of financial wealth. A comparison of the Fraser Institute’s ranking of countries according to “economic freedom,” with the United Nations’ Human Development Index (which ranks countries according to concrete measures of living standards such as income levels, life expectancy, and education) shows no correlation between the two.4 Advocates of laissez faire 2 economic policies argue that these policies will enhance economic growth. This is not clear, nor is it even clear whether faster growth under a laissez faire regime translates in any predictable way into higher living standards for the majority of the subjects of that regime. Even on its own terms, the pro-wealth bias of the Fraser Institute approach is clear. For example, one key feature of freedom, Fraser-style, is low taxes. This is an obviously controversial assumption in and of itself: a different vision of an economically free society might well feature the provision of essential services (such as health care or education) on a noncommercial basis by government or some other collective agency, and this would naturally require the collection of taxes to support those services. But even if one accepts the starting assumption that low taxes equal more freedom, it is puzzling that the Fraser Institute captures this effect only by examining the marginal tax rate paid by the highest-income taxpayers. A jurisdiction could impose relatively high taxes on the bottom 90 percent of its population, but relatively low marginal rates for the richest 10 percent, yet would still be considered relatively “free.”5 Why would the “distortionary” or “discriminatory” effects of high taxes on the bottom 90 percent of society not be of even more interest than the taxes paid by the highest-income taxpayers? Clearly, in the Fraser Institute’s approach, it is economic freedom for the wealthy and powerful that really matters. It is tempting but incorrect to identify “economic freedom,” according to the Fraser Institute’s view, with “less government”. Obviously the Fraser Institute’s approach favours an absence of government intervention in key areas (such as labour markets, international trade, economic regulation, and so on). But this does not imply that government is absent or powerless. Indeed, economic freedom for the wealthy and powerful presupposes a strong and interventionist government, one which is prepared to intervene forcefully to protect the interests of those who own financial wealth or operate businesses. For example, the protection of private property features centrally in this conception of freedom. In modern times, this means more than simply providing the legal and enforcement apparatus to protect the haves from the have-nots. In an era of computer software and biotechnology, the very concept of “private property” has changed dramatically to include knowledge as well as commodities and money. This has required ever-more forceful and creative interventions by government to protect the intellectual property rights of firms and investors.6 Similarly, the Fraser Institute’s emphasis on the maintenance of low inflation provides another example of the still-powerful but increasingly one-sided nature of government in an economically “free” society. Standard economic theory holds that to keep inflation at a low, steady level, the central bank of an economy must intervene forcefully with higher interest rates to slow down job-creation should the unemployment rate fall close to its long-run “natural” level (technically defined as the Non-Accelerating Inflation Rate of Unemployment, or NAIRU). What an incredibly powerful lever of government intrusion into the economy! A single unelected government agency is granted the power to slow down the expansion of the entire country, and fundamentally shift relative prices (by increasing the price of capital relative to the prices of other factors), in order to preserve a low-inflation economic environment. The costs of this policy are 3 huge, borne not just by the unemployed but by all workers (whose security and bargaining power are undermined by the deliberate maintenance of permanent unemployment). Yet this is a form of government intervention that, in the Fraser Institute model, promotes economic freedom—more specifically, by cementing the freedom of the owners of financial wealth from the risk that their wealth will be inflated away. Despite its one-sided and ideologically biased approach, the Fraser Institute’s index of economic freedom is not uninformative. It is an interesting example of an attempt to develop a quantitative index on the basis of numerous underlying components, many of which are qualitative in nature. And the final scores provide a useful measure of international variation in economic and institutional arrangements. But this variation does not reflect differences in “economic freedom” defined in any broader, genuine way. The index should be renamed an Index of Capitalist Freedom. To what extent are a jurisdiction’s policies and institutions designed so as to maximize the ability of business owners and financial investors to extract maximum profit on a sustained basis from their undertakings? The Fraser Institute’s ranking project constitutes a good first start at answering this question. In 1998 the Fraser Institute announced that it was extending its economic freedom project to include the development of an index comparing economic freedom across the ten Canadian provinces. The motivation for this work is clearly as much political as it is academic. Weeks before the study was even released, an impatient Fraser Institute director Michael Walker leaked its key finding in a newspaper column: British Columbia ranked very low on the freedom index, while Alberta and Ontario vied for top spot. Indeed, embarrassing the beleaguered NDP government in B.C. seems to be a key motivation for the entire project. “While Alberta and Ontario vie for top spot among the provinces, British Columbia has slowly, deliberately and as a direct result of changes in policy fallen to the seventh-freest province,” crowed Walker.7 ‘Rightwing good, left-wing bad’ seems to be a simple mnemonic that will help readers grasp the essential thrust of this Animal-Farm-like analysis of economic policy in Canada. The Fraser Institute’s inter-provincial rankings tell us a lot about the relative pro-business orientation of differing provincial governments, but it does not tell us much about the economic freedom of most Canadians. Economic Freedom (for the Rest of Us) If one is not a capitalist, then an Index of Capitalist Freedom will not provide much indication of one’s economic rights, protections, and prospects. Most people in the world are not capitalists, in the sense that they do not own businesses or wealth on the basis of which they can support themselves. We need an alternative index, therefore, to measure the extent to which differing policies and institutions make it easier or more difficult for us to freely and productively lead our economic lives. This paper presents an index of Economic Freedom (for the Rest of Us) Index, called 4 the EFRU index, which is intended to measure the impact of government policies, rules, and institutions on the economic prospects of those members of society who must work for a living. The index is calculated on the basis of 13 different component variables (some of which are themselves composite in nature, so that a total of some 20 different economic variables is considered by the analysis).8 These variables are in turn grouped and weighted into three broad categories, each of which composes one-third of the final index: 1) 2) 3) Employment (representing freedom from exclusion) Earnings (representing freedom from deprivation) Equality and Security (representing freedom from discrimination and fear) Together these categories reflect the extent to which the members of society are able to support themselves in productive employment, receive an income which reflects their productivity and allows for an adequate standard of living, and are protected from arbitrary discrimination or dislocation on the basis of personal characteristics or economic misfortune. The individual variables contained within each of these three broad categories, and their weighting within those categories, are summarized as follows: Employment: C C C employment rate (2/3) index of underemployment (part-time and self-employment; 1/6) duration of unemployment (1/6) Earnings: C C C average weekly wages and salaries (1/2) average earnings as a share of productivity per worker (1/4) minimum wage (1/4) Equality and Security: C C C C C C C equality of income distribution (1/6) earnings gap between male and female workers (1/6) earnings gap between hourly employees and salaried workers (1/6) family poverty rate (1/8) government spending on public programs per capita (1/8) social assistance benefits (1/8) average job tenure (1/8) Details on the choice and definition of variables, their weighting, sources of data, and the quantitative methodology followed in the construction of the final EFRU scores are provided in 5 the Appendix. The selection of variables appearing in the EFRU index differs notably from the Fraser Institute’s ranking (which did not consider the employability or earning power of workers to be a relevant issue). A few variables might actually appear in both indices, but with opposite signs. For example, the Fraser Institute considers public spending to be an intrusion on economic freedom, while the present study recognizes the beneficial impact of public services in underwriting the standard of living of working people and thus enhancing their bargaining power in the labour market. Similarly, the Fraser Institute would consider a high minimum wage to be a violation of the right of employers to hire labour services at the lowest price that can be extracted on the basis of compulsion and hardship prevailing in the labour market at any point in time. In our case, however, a higher minimum wage clearly enhances the freedom of low-wage workers: they are guaranteed a more adequate minimum compensation in return for their work effort.9 Once again, economic freedom is very much in the eye of the beholder. Any attempt to construct a quantitative index on the basis of numerous weighted composite variables is to a large extent inherently arbitrary, and the selection of variables (and their weighting) for the EFRU index will undoubtedly spark some disagreement. Future research in this field will hopefully allow for the inclusion of additional variables. For example, the entire subject of labour and collective bargaining freedoms—so crucial to the economic prospects of many working people—is not included in the current index, due both to data availability and to conceptual problems.10 Still, the present index is less arbitrary, in some important ways, than the Fraser Institute’s economic freedom index—which not only assigns quantitative weightings to component variables in a necessarily subjective manner, but even incorporates some series as index components which themselves consist of scores subjectively assigned to reflect variance in some qualitative indicator.11 In contrast, all of the data utilized in computing the EFRU index are publicly available series collected by statistical agencies following normal survey methodologies. Finally, the present study is released in the context of more significant longer-term work by other researchers to develop alternative “quality of life” indicators.12 Together these initiatives hold great potential for developing a better understanding of the real determinants of living standards and how those standards are changing over time. The particular focus of the present study is to engage explicitly with the conception of “economic freedom” that is advanced by the advocates of free-market economic policies. EFRU Scores for the Canadian Provinces The EFRU index is composed of very different component variables than a marketoriented economic freedom index. It is to be expected, therefore, that the scores and rankings produced by the EFRU approach will differ considerably from those of the Fraser Institute’s measurement tool—although the EFRU results do contain some surprises.13 A summary of the EFRU scores and rankings for the ten provinces for 1990 and 1997 is provided in Table 1. A 6 more detailed breakdown of EFRU scores between the three major components of the index is provided in Table 2.14 Complete listings of both raw data and EFRU scores for each of the 13 component series for each province are provided in the Appendix. Ontario is found to exhibit the highest level of economic freedom for workers of the ten provinces, on the strength of the highest earnings in Canada, and a relatively good employment record. Ontario scores relatively lower (seventh out of ten) on the equality and security index, which pulls down its overall score somewhat. Second place is held by British Columbia, which ties Ontario with the highest earnings, exhibits an average score on employment, but ranks poorly in terms of equality and security (due largely to the most unequal distribution of income in Canada). Quebec places surprisingly high in the EFRU ranking: third. Despite a weaker economy and poor job-creation (Quebec ranks eighth out of ten on the employment index), Quebec scores near the top of both the earnings and the equality/security indices and this pulls up its overall ranking. Through its wage and social market policies, Quebec seems to do a better job of transmitting a limited amount of economic growth and job creation into income and security for its population. At the other extreme, Alberta’s overall EFRU score is mediocre—it ties for fifth place with Prince Edward Island—despite its booming economy and a first-place ranking in the employment index. Growth and job-creation in Alberta have been strong, yet the province’s labour market and social policies are such that this growth has not translated into concrete gains for most workers. Lowly P.E.I., meanwhile, tops the nation in terms of the Equality and Security index (thanks especially to the lowest poverty rate and second-most equal distribution of income), and this explains its relatively high placement in the overall EFRU ranking. Hard-hit by the collapse of the fishing industry and deep reductions in government social programs, the other Atlantic provinces (especially Newfoundland and Nova Scotia) fill the bottom positions in the EFRU ranking. As interesting as the current EFRU scores attained by each province are the directions and magnitudes of change in economic freedom. To develop a sense of the evolution of economic freedom in Canada through the restructuring and stagnation which has characterized our national economy for most of this period, EFRU scores were calculated for each province for 1990, using the same methodology.15 As expected, overall economic freedom in Canada has deteriorated during the lean 1990s (by about 2 percentage points on the arbitrary EFRU scale). But it is perhaps surprising how relatively stable the EFRU scores have remained, despite the economic turbulence of the decade (average Canada-wide EFRU scores for each of the component series for 1990 and 1997 are presented in Table 3) . In particular, there was virtually no change whatsoever in the Earnings or Equality and Security sub-indices, which both retained their value throughout the decade (the Earnings index fell marginally through the 1990s, while the Equality and Security index actually rose marginally). Thus all of the decline in overall EFRU scores in Canada is attributable to the deterioration in the Employment category—driven in particular by a decline in the overall employment rate from about 62 percent of working age adults in 1990 to 59 percent in 1997. In terms of individual provinces, the decline in economic freedom during the 1990s was 7 experienced quite broadly, with overall EFRU scores falling for seven of the ten provinces. The hard-hit Atlantic provinces led the decline, with Newfoundland’s score dropping by almost 7 points, and Nova Scotia’s by close to 4 points. Interestingly, Ontario experienced the third largest deterioration in economic freedom in the 1990s, with an overall decline of close to 3 points; a steep fall in the province’s employment record (the Employment EFRU for Ontario fell by 12 points between 1990 and 1997), and the erosion of social programs in the province, account for the decline. In contrast, British Columbia experienced the largest increase in its overall EFRU score during the same period, by 1.6 points. The ranking of the top three provinces was unchanged during the 1990s—Ontario first, B.C. second, and Quebec third—but the gap between Ontario and B.C. closed from over 5 points in 1990 to just a single point in 1997. If current trends continue, B.C. could soon challenge Ontario for the highest EFRU ranking in the country. No-one should conclude, therefore, that Ontario’s first-place ranking in the EFRU index somehow constitutes an endorsation of the policies followed by the current Ontario government. Quite to the contrary, under its current government, Ontario’s long-standing position as a leader in terms of rights and freedoms for workers has been placed very much in jeopardy. Two other provinces—Quebec and Manitoba—also experienced improvements in economic freedom, thanks to stronger earnings and a better record of equality and security in both provinces. Lessons and Conclusions Economic freedom is in the eye of the beholder, and the dramatic changes which have been experienced in Canada’s economy during the 1990s provide ample evidence of this edict. Economic policies in Canada have shifted substantially in the direction of deregulation, a smaller and less interventionist government, and greater reliance on private markets as the guiding force of economic and social development. According to the perspective of free-market economics, therefore, Canada’s has become a significantly freer economy.16 Yet for most Canadians—those who must work in a paying job to support themselves, rather than skimming the income from accumulated wealth or privately-owned businesses—economic freedom has declined. Thanks in large part to those same conservative economic policies—tight-money anti-inflation policy, radical cutbacks in government spending, and the erosion of institutional protections in labour markets and social policies—economic life for most Canadians has become more challenging, and indeed oppressive. Even for those who have been critical of the free-market drift of economic policy in Canada, however, the results presented here contain some important lessons and surprises. Here are a few insights that can be gleaned from the evolution of the EFRU index in Canada over the 1990s: i. The Social Safety Net Matters a Lot: If anything, the decline in average EFRU scores in Canada during the 1990s is surprising for its modest size. Economic times have been tough; labour markets have been chronically 8 (some would say deliberately) slack. Yet average earnings have not declined, and according to a range of social and economic indices, even the level of social equality has not yet been definitively undermined.17 This may, to some extent, simply reflect a long lag time with which changes in the deep institutional structure of the economy affect the concrete living conditions of the bulk of the population, but it also reflects the important moderating impact of Canada’s remaining social programs in alleviating the impact of the sustained downturn on the living standards of Canadians. A similar phenomenon is also visible across the provinces. The four Atlantic provinces which generally rank poorly in terms of employment and earnings all register above-average scores on the Equality and Security sub-index which pull up the overall rankings of these hard-hit provinces. In short, the EFRU evidence suggests a strong and lasting influence of those progressive institutional structures (unions, minimum wages, progressive income taxes, social programs) which are clearly moderating the otherwise negative impact of market forces on the economic freedom of Canadians. This is not to suggest that we should be sanguine about the direction in which Canadian society is evolving, or that the struggle to reverse the conservative policy changes of the past decade should be abated. But it would enhance the confidence and bargaining position of the advocates of greater equality and opportunity if we better understood just how strong Canada’s system of social and labour policies has proven to be. Despite the most radical and sustained rightward shift in the history of Canadian economic policy, the earning power and social security of Canadians has not (so far) been sacrificed. ii. Politics Don’t Matter as Much as You’d Think: By focusing on the philosophical orientation of economic policy, rather than on the concrete living conditions of the citizenry, the Fraser Institute’s approach to measuring economic freedom tends to quickly and predictably reflect changes in the political outlook of ruling governments. A right-wing government which cuts taxes and privatizes Crown corporations will move quickly up the freedom ranking, which consequently tends to be headed by jurisdictions rules by conservative political parties. In contrast, the link between the politics of the day and economic freedom reflected in the EFRU index is less direct and predictable, and for this reason EFRU scores provide more informative and reliable insight into the true differences in deeper underlying structural relationships which impact upon the economic freedom of the majority of Canadians. It may seem strange that an index designed by a union economist should find that economic freedom in Mike Harris’ Ontario exceeds that of any other province in Canada. But when the concrete reality of Ontario’s economy (rather than shorter-term political rhetoric) is examined, we find many lasting positive features: a diversified economy, high wages, strong unions, and a social safety net that despite recent cutbacks still ranks as “average.” Obviously, more progressive and humane policies in Ontario could generate a higher degree of economic freedom for its population. But critics of neoconservative policies can take heart that even a radically conservative government (after three years, anyway) has not been able to dismantle 9 many of the key structural and institutional features that still make Ontario a relatively good place for workers. To put it bluntly, electing right-wing Conservatives is no guarantee that economic freedom will evaporate, not does electing the NDP guarantee that economic freedom will improve. True economic freedom for workers depends on a complex and slowly-changing mix of economic development, social policy, and institutional struggle. By keeping their eyes on this longer-term process, progressives may be less likely to become unduly pessimistic in the wake of particular conservative policy victories, nor so unjustifiably optimistic in the wake of particular progressive electoral successes. iii. Job-Creation is a Pillar of Economic Freedom By any calculation, employment opportunities are a crucial determinant of the economic freedom of working people. A jurisdiction which featured progressive social and labour market policies, but a weak job-creation record, would experience only uneven progress toward the goal of enhanced freedom for its working population.18 The longer-run effort to enhance economic freedom therefore requires the development of an effective and convincing alternative jobcreation strategy. Such a strategy could not rely on the tax cuts and regressive labour market policies which have helped to create some jobs in jurisdictions such as Ontario or Alberta, since these changes undermine economic freedom even as employment grows. The development of a full-employment strategy that is somehow and somewhat independent of the profit-seeking imperatives of private business, and hence able to create jobs without undermining progressive labour market and social policies, remains a crucial challenge for progressives in Canada and elsewhere. 10 Table 1 Economic Freedom (for the Rest of Us) Ranking by Province 1997 Rank 1997 Score Change since 1990 1990 Rank Ontario 1 105.8 -2.7 points 1 British Columbia 2 104.6 +1.6 points 2 Quebec 3 102.4 +0.8 points 3 Manitoba 4 100.6 +0.2 points 9 Alberta 5 99.4 -1.5 points 6 P.E.I. 5 99.4 -2.1 points 4 Saskatchewan 7 98.2 -2.4 points 8 New Brunswick 7 98.2 -0.9 points 10 Nova Scotia 9 97.4 -3.6 points 5 Newfoundland 10 94.0 -6.8 points 7 SOURCE: Author’s calculations as described in text. 11 Table 2 EFRU Rankings by Category, 1997 Overall Rank Employment Score Rank Score Rank Score Rank Ontario 1 102.7 4 115.2 1 99.7 7 British Columbia 2 100.3 6 115.2 1 98.2 9 Quebec 3 97.0 8 107.5 3 102.7 3 Manitoba 4 105.8 2 95.9 6 100.0 6 Alberta 5 111.6 1 96.7 4 90.0 10 P.E.I. 5 101.4 5 92.1 10 104.7 1 Saskatchewan 7 103.1 3 93.0 9 98.7 8 New Brunswick 7 98.1 7 96.4 5 100.1 5 Nova Scotia 9 95.7 9 94.2 7 102.4 4 Newfoundland 10 84.5 10 93.9 8 103.6 2 SOURCE: Author’s calculations as described in text. 12 Earnings Equality/Security Table 3 Change in Economic Freedom (for the Rest of Us) Canada, 1990-1997, by Category Weighting1 1997 Avg. 1990 Avg. Change EMPLOYMENT SUB-INDEX 1/3 100 105.7 -5.7 pts. Employment Rate 2/3 100 102.9 -2.9 pts. Underemployment Index 1/6 100 106.5 -6.5 pts. Duration of Unemployment 1/6 100 116.0 -16.0 pts. 1/3 100 100.1 -0.1 pts. Weekly Earnings 1/2 100 99.4 +0.6 pts. Earnings / Productivity 1/4 100 106.8 -6.8 pts. Minimum Wage 1/4 100 94.8 +5.2 pts. EQUALITY/SECURITY SUB-INDEX 1/3 100 99.4 +0.6 pts. Income Distribution 1/6 100 101.0 -1.0 pts. Gender Equality 1/6 100 92.8 +7.2 pts. Hourly/Salaried Equality 1/6 100 98.9 +1.1 pts. Poverty Rate 1/8 100 109.3 -9.3 pts. Public Program Spending 1/8 100 100.8 -0.8 pts. Social Assistance Rate 1/8 100 104.2 -4.2 pts. Job Tenure 1/8 100 90.7 +9.3 pts. EARNINGS SUB-INDEX SOURCE: Author’s calculations as described in text. 1. Sub-indices each compose 1/3 of final EFRU index. Indicated weightings for individual components indicate their weighting within their respective sub-index. 13 Appendix: Methodology and Data The EFRU (Economic Freedom for the Rest of Us) index assigns a numerical score to each jurisdiction on the basis of a weighted composite of 13 different indicators of the economic freedom of Canadians who support themselves through real work activity (rather than through income derived from financial wealth). The index is arbitrarily scaled so that the average EFRU score for Canada as a whole19 in 1997 equaled 100. However, the scores are cardinal (not just ordinal), in the sense that the relative positioning (not just the rank ordering) of differing jurisdictions is indicated by their relative scores. This methodology requires that scores on all statistical components entering the EFRU be converted into an index with mean equal to 100, such that higher numbers signify “more freedom.”20 To further ensure that each component of the EFRU exerts its appropriate weight in the final index calculation and provincial ranking, the component series are further normalized so that their standard deviation (a measure of the relative dispersal of individual province scores around the Canada average) equals 10.21 The procedure followed in this normalization process is as follows. First all individual scores in a series were multiplied by the ratio of 100 over the sample mean. Then the deviation between each adjusted individual score and the adjusted sample mean (100) was normalized by multiplying by the ratio of 10 over the original sample standard deviation. For inverted series, the sign of the adjusted deviation from 100 was also reversed. The final normalized scores were then calculated by adding the normalized deviation to the adjusted sample mean. Data were available for most component series of the EFRU index for 1997. In a few cases, however, 1996 data were the most recent available; these series (all of which enter into the Equality and Security sub-index) are noted in the detailed descriptions which follow. The EFRU index is composed of three equally-weighted broad categories, each of which composes one-third of the final index: 1) 2) 3) Employment (freedom from exclusion) Earnings (freedom from deprivation) Equality and Security (freedom from discrimination and fear) Each of the four broad category scores is in turn computed on the basis of weighted inputs of a variety of different sub-components, as specified below: 1) Employment: No single factor is more important to the ability of a jurisdiction’s citizens to live healthy, meaningful lives than the opportunity to find productive employment. In this category we consider the employment rate as a general indicator both of general labour market conditions and 14 of the relative ability of individuals to participate in the labour market. The employment rate has become a more meaningful indicator of labour market conditions in recent years than the unemployment rate, since the latter is subject to variation as a result of changing patterns of labour force participation. The employment rate constitutes two-thirds of the total employment index. The employment index also incorporates a composite measure of underemployment, to capture the fact that the simple distinction between employed and unemployed will fail to capture the underemployment of many Canadians are not meaningfully employed to the full extent actually desired. This measure is calculated on the basis of the proportion of total employed workers who are part-time and self-employed. Given the preponderance of involuntary part-time work and marginal self-employment activity by workers forced from paying jobs, this measure will provide an approximate indicator of the extent to which simple employment statistics overstate the opportunities for real productive work Canadians.22 Finally, the employment index also reflects the average duration of unemployment in each province, measured in weeks. The effect of varying levels of unemployment from province to province is captured via the inclusion of the employment rate. However, varying durations of unemployment (capturing the typical length of time it takes an unemployed worker to find a new position) reflects an additional and separate dimension of the unemployment problem. A province may have a high unemployment rate but a short unemployment duration (such as P.E.I.), which suggests that the burden of unemployment in that province is somewhat more equally shared (between a larger number of job-seekers who are each unemployed for a relatively shorter period of time). To sum up, the employment component of the EFRU index is composed of the following variables: 2) C Employment rate (employment as a share of working age population; composes 2/3 of Employment score; source Statistics Canada Catalogue 71-201, Historical Labour Force Statistics). C Index of underemployment (1 less the sum of part-time and self-employment as a share of total employment; composes 1/6 of Employment score; source Statistics Canada Catalogue 71-201, Historical Labour Force Statistics, and unpublished Statistics Canada labour force data). C Average duration of unemployment (in weeks; inverted; composes 1/6 of Employment score; source unpublished Statistics Canada labour force data). Earnings: Most Canadians must not only be employed in order to support themselves, they must 15 earn a decent income in return for that work that reflects both the absolute costs of supporting oneself and one’s family, and also the general level of productivity of the work itself. For this reason we consider both the absolute level of labour income (measured by the industrial composite of average weekly wages and salaries), and employees’ relative share of the economic pie in their jurisdiction (captured by the share of earnings in the average GDP produced per worker). This dual approach is adopted because we do not wish to completely equate a “free” jurisdiction with a “rich” one; a free but less economically developed jurisdiction may pay lower wages in absolute terms, even though its working citizens may actually be receiving a higher share of total output. Nevertheless, due to the importance of living standards in absolute terms to the freedom of workers from poverty, the absolute level of earnings is ascribed primary weighting. The Earnings sub-index also includes the minimum wage prevailing in each province, as a measure of the relative economic freedom of the lowest-paid members of the workforce. To sum up, the Earnings score includes the following variables: 3) C Average weekly earnings (dollars per week, salaried and hourly; composes 1/2 of Earnings score; source Statistics Canada Catalogue 72-002, Annual Estimates of Employment, Earnings and Hours). C Average weekly earnings as a share of total provincial GDP at market prices per employee (percent; composes 1/4 of Earnings score; source Statistics Canada Catalogue 13-213, Provincial Economic Accounts, Annual Estimates, and previously cited labour market data). C Provincial statutory minimum wage prevailing at December 31 (dollars per hour; composes 1/4 of Earnings score; source Human Resources Canada, Workplace Information Directorate, unpublished data). Equality and Security: The members of an economically free society will be compensated without regard to personal gender, race, or socio-economic features over which they have no control; they will be free from discrimination, in this broad sense. Similarly, the scale of income differences between individuals resulting from differences in occupation will be moderated. Finally, a free society will provide for the economic well-being of its citizens on a consistent and ongoing basis, so that this well-being is not unduly put at risk by negative events (whether due to broader economic developments or unfortunate personal circumstance). In terms of economic equality, the Equality and Security index considers the relative distribution of total income between high-income and low-income families, captured by estimated provincial Gini coefficients.23 It also includes the relative distribution of income between hourly employees (presumably those working in less-skilled or less senior positions) and 16 salaried professions. Finally, it also includes a measure of the male-female earnings gap as an indicator of the relative freedom from discrimination in the economy. Unfortunately, comparable data on income differentials according to race are not available; it is reasonably safe to assume, however, that an economy which is relatively free on these other three criteria will tend to exhibit the same sorts of legal and social policy patterns that would ensure greater freedom from racial economic discrimination. We capture the stability and security of workers’ living standards through four different indices. The family poverty rate measures the proportion of families whose incomes fall below a relative low-income cut-off. We proxy the level of provision of basic social and human services (which underpin the overall standard of living of most citizens) by government with the level of program spending by all levels of government per capita.24 This measure might be considered the “social wage” of a society. The relative degree of protection for families who fall through the cracks of our volatile economy is captured by including provincial social assistance rates. Finally, a measure of the degree of turnover and general instability in the labour market is provided by the average job tenure of employed workers in each province. To sum up, the Equality and Security sub-index of the EFRU includes the following seven components: C After-tax Gini coefficient for all household units (absolute number between zero and one; inverted; 1996 data; composes 1/6 of Equality and Security score; source Statistics Canada Catalogue 13-210, Income After Tax, Distributions by Size in Canada). C Ratio of female average annual earnings to male average weekly earnings (percent; 1996 data; composes 1/6 of Equality and Security score; source Statistics Canada Catalogue 13-217, Earnings of Men and Women). C Ratio of average hourly wage times average weekly hours of work for hourly employees to average weekly earnings for all hourly and salaried employees (percent; composes 1/6 of Equality score; source Statistics Canada Catalogue 72002, Annual Estimates of Employment, Earnings and Hours). C Family poverty rate (percent of families below low-income threshold; inverted; 1996 data; composes 1/8 of Equality and Security score; source Statistics Canada Catalogue 13-207, Income Distributions by Size in Canada). C Total government program spending per capita (dollars; 1996 data; composes 1/8 of Equality and Security score; source Statistics Canada Catalogue 13-213, Provincial Economic Accounts, Annual Estimates). C Total social assistance income for a single-parent family with one child (1996 17 dollar terms; 1996 data; composes 1/8 of Equality and Security score; source National Council of Welfare, Welfare Incomes 1996). C Average job tenure of employed workers (weeks; composes 1/8 of Equality and Security score; source unpublished Statistics Canada labour force data). Note also that the Equality and Security sub-index can be decomposed into two equal component parts, an Equality index (consisting of the first three measures, all dealing with income distribution) and a Security index (consisting of the latter four measures, all of which reflect various dimensions of social security). Intertemporal Comparisons The goal of the present study is not solely to compare the level of economic freedom across provinces; we are also interested in how that level has changed over time. To provide an indication of the impact of economic restructuring and the prolonged recession of the 1990s on economic freedom in Canada, data on all component series were also assembled for 1990.25 EFRU scores and corresponding provincial rankings could then be generated with this earlier data set. To ensure that the 1990 EFRU scores could be compared to the 1997 scores, the following methodology was adopted. Most of the 13 component series of the EFRU index are measured in terms that are not dependent on the nominal level of prices in the economy. In three cases, however, nominal magnitudes are utilized: weekly earnings, the hourly minimum wage, and total government program spending per capita. To ensure comparability with the 1997 data, therefore, these series had to be converted into real 1997-dollar terms, by deflating by the proportional rise in consumer prices (measured by the rise in the CPI) between 1990 and 1997.26 The procedure through which the raw data series were normalized to calculate the 1997 EFRU scores would also distort the comparability of EFRU scores over time. Recall that all 13 component data series for 1997 were normalized such that their mean equals 100, their standard deviation equals 10, and a higher number corresponds to greater freedom. If the same normalization were performed on the 1990 data, then no intertemporal change in EFRU scores would be visible (since the average score in both years would equal 100 by design). Thus the 1990 data were normalized with the same absolute factors as were utilized in the 1997 normalization.27 The means and standard deviations of these adjusted 1990 series therefore differ from 100 and 10, respectively; when the normalized mean score of a 1990 series exceeds 100, this implies that this indicator of economic freedom has deteriorated over the 1990s, and if the adjusted standard deviation exceeds 10, this implies that provincial scores have become more dispersed during the 1990s. Notes 18 1. “Free Canadian savings to cross the border,” Globe and Mail, September 19, 1997. 2. In this context, the contrast between so-called economic freedoms and traditional politicallyrooted conceptions of freedom is often striking. The Fraser Institute’s economic freedom project was explicitly motivated by a frustration that traditional conceptions of democracy downgraded the importance of private economic rights; see Michael Walker, “The historical development of the economic freedom index,” in James Gwartney, Robert Lawson and Walter Block, eds., Economic Freedom of the World 1975-1995 (Vancouver: Fraser Institute, 1996). Thanks to this approach, countries in which the guarantee of basic democratic rights is questionable at best (such as Thailand, South Korea, or Guatemala) can nevertheless rank highly in terms of “economic freedom.” See Murray Dobbin, The Myth of the Good Corporate Citizen (Ottawa: Canadian Centre for Policy Alternatives, 1998) for a more detailed critique of the anti-democratic bent of the economic freedom meter. 3. See James Gwartney, Robert Lawson and Walter Block, eds., Economic Freedom of the World 1975-1995 (Vancouver: Fraser Institute, 1996), and subsequent editions. 4. In fact, the correlation coefficient between the top 20 countries on the UN’s list and their corresponding Fraser Institute scores for 1995 was slightly negative—implying a small negative relationship between the degree of “economic freedom” of a society and the actual living conditions of its population. See Jim Stanford, “Economic freedom,” Globe and Mail, January 26 1996. 5. This is not a hypothetical issue. Tax policies vary across jurisdictions not solely according to the level of taxes collected, but also according to the relative progressivity of the collection systems (that is, the extent to which higher-income taxpayers bear a relatively higher tax burden). Alberta has long boasted that it has the lowest income taxes in Canada, and the usual way of demonstrating this is by comparing top marginal tax rates across provinces. For high-income earners, income taxes are indeed lowest in Alberta; but for a single taxpayer with one dependent and annual income of $20,000, Alberta’s taxes are the fifth highest in Canada, and hence are relatively “un-free” (calculated from Canadian Tax Foundation, Finances of the Nation 1996, Table 3.18). Whether Alberta’s tax structure promotes “freedom,” therefore, depends entirely on how much money one makes. 6. Modern free trade treaties, for example, typically include more text detailing the new regulatory powers of governments to hunt down and prosecute intellectual property violators, than describing the elimination of tariffs and other unwanted regulatory powers. It is not clear at all that government is getting smaller or less intrusive, even as economic “freedom” (in the realm of international trade) is being consolidated. 7. “Clark closes ears to warnings from industry, investors,” Financial Post, July 18 1998. 8. The EFRU index is thus comparable in its empirical complexity to the Fraser Institute’s interprovincial freedom index, which includes 12 components. 19 9. This is not the place to review the long-standing debate over whether minimum wages increase unemployment; suffice it to report that most modern economic analysis of this question has found a negligible or even positive impact of minimum wages on low-wage employment. For a summary of recent evidence, see David Card and Alan B. Krueger, Myth and Measurement: The New Economics of the Minimum Wage (Princeton: Princeton University Press, 1995); for a Canadian discussion, see Errol Black and Lisa Shaw, The Case for a Strong Minimum Wage Policy (Ottawa: Canadian Centre for Policy Alternatives, 1998). 10. Surely the right to organize a union and bargain collectively is a central economic freedom. It is not entirely clear, however, how this freedom should be measured. Some studies have utilized the union penetration rate as a proxy for collective bargaining freedoms, but this potentially confuses the freedom to join a union with the decision to actually join one. Future research and discussion on this issue is welcomed. 11. For example, the Fraser Institute global freedom index includes a measure of the “freedom to enter and compete in markets,” varying between 0 and 10, which is constructed on the basis of a subjective evaluation of the regulatory policies of different countries. Several other component series are similarly and subjectively constructed. 12. See, for notable Canadian examples, Lars Osberg and Andrew Sharpe, “An index of economic well-being for Canada,” Centre for the Study of Living Standards, May 1998; David Cosby, “A genuine progress indicator for Canada: an alternative to growth as a measure of progress,” in Bruce Campbell and John Loxley, eds., The Alternative Federal Budget Papers 1997 (Ottawa: Canadian Centre for Policy Alternatives, 1997); and “Measuring well-being: proceedings from a symposium on social indicators,” Canadian Council on Social Development, November 1996. 13. In this context, questions are raised regarding Canada’s consistent top ranking in the United Nation’s Human Development Index and other international comparison projects. Were an international comparison to be conducted that included the various socio-economic variables considered by the EFRU index, Canada’s performance would not likely be so praiseworthy. 14. As explained in detail in the Appendix, the EFRU scores for each component series (as well as the overall scores) are normalized so that an unweighted average of the ten provinces for 1997 equals 100, and so that a higher score implies more freedom. 15. The same absolute normalization factors were applied in the 1990 calculations as were utilized in the 1997 calculation; thus the 1990 scores are expressed relative to a 1997 Canada-wide average within each component series equal to 100. Any score higher in 1990 than the corresponding score for 1997 implies that economic freedom has deteriorated during that period. See Appendix for more details. 16. Even the Fraser Institute verifies this finding by virtue of Canada’s consistently high rankings in its global economic freedom index. Of course, at any point in time the Fraser Institute would 20 always prefer policy to be more laissez faire than it is. But even by their standards Canada possesses one of the most deregulated, business-oriented economies in the world. 17. Particularly striking in this regard has been the near-stability of the after-tax distribution of income in Canada, despite the growing polarization of “market” income between low- and highincome households. 18. For example, though B.C. and Quebec scored highly in the EFRU index on the strength of progressive labour market and social policies, scores in both provinces were considerably undermined by less favourable job-creation records. 19. The Canada-wide average in each case is an unweighted mean of the ten provincial scores. 20.This approach requires that some of the raw data entering the EFRU index calculation be inverted so that a higher absolute value indicates more freedom; the inverted series are identified in the detailed description below. 21. Without this additional normalization, series which exhibited a larger degree of interprovincial variation in relative arithmetic terms (such as, for example, the average duration of unemployment) would effectively carry more than their assigned weight in the final index, while those with relatively less relative variation (such as the Gini coefficients of income distribution) would carry less influence than intended. The normalization of standard deviations ensures that the adjusted values for all series will approximately vary across provinces between 90 and 110. 22. To be sure, many Canadians work part-time or are self-employed by choice; moreover, since some Canadians will be both part-time workers and self-employed, the index of underemployment as presently constructed will involve a limited degree of double-counting. Nevertheless, since there is widespread agreement that the significant rise in both part-time employment and self-employment in recent years largely reflects sustained weakness in the paid labour market, this index still serves as a useful proxy for the level of precariousness in employment relationships. 23. The Gini coefficient is a composite statistical measure of income distribution between households, in which a higher score corresponds to greater inequality. A Gini score of 0 corresponds to a perfectly equal distribution (in which all households receive the same income), while a score of 1 indicates perfect inequality (in which a single household receives all the income produced in the society). 24. It would be preferable, of course, to focus on particular types of program spending which are considered to be most important in underpinning living standards (presumably including important programs such as health care and education, while excluding line items such as military spending). The availability of comparable and timely data across jurisdictions, however, prohibits this more targeted approach; a task for future research into quality of life indicators would be to improve the accuracy of this measure. 21 25. For those series for which 1996 data were the most recent available, 1989 data were used in the calculation of EFRU scores for 1990. 26. Since the data on program spending per capita were lagged one year, the relevant deflator in this case is the rise in the CPI between 1989 and 1996. 27. Specifically, the original 1990 scores were multiplied by the sample mean of the corresponding series for 1997. Then the deviations between these adjusted scores and their new mean were multiplied by the ratio of 10 over the actual standard deviation of the original 1997 series; for inverted series, the sign of the deviation is also reversed. 22